Contributions published at Financial Engineering (Markus Leippold)

Large amounts of statistical significant predictors in asset pricing make testing new factors chal-lenging. The aim of this paper is to adopt the post-double-selection LASSO and use it to test the significance of uncertainty indices in asset pricing. Our test shows that all the uncertainty factors are statistically not significant in predicting cross-section of expected returns. In addition, we found out that the post-double-selection LASSO does differentiate with post-single-selection LASSO. However, the result with different missing values imputations and test assets are not stable.

null

null

This paper presents a process which allows for stochastically scheduled events. To retain its tract-ability, we derive its characteristic function with arbitrary delay assumptions. Moreover, we develop an estimation strategy to extract information of model parameters as well as scheduled events from SPX option data. Specifically, we apply di↵erent approaches, e.g. least squares fitting and particle filter, to an model-implied quantity-VIX2 to estimate the model parameters. Meanwhile, we can identify the event jump times by looking at the VIX2 term structure. Keywords: Scheduled events; Characteristic function; SPX option; VIX; Particle filter

null

We present SUMO, a neural attention-based approach that learns to establish the correctness of textual claims based on evidence in the form of text documents (e.g., news articles or Web documents). SUMO further generates an extractive summary by presenting a diversified set of sentences from the documents that explain its decision on the correctness of the textual claim. Prior approaches to address the problem of fact checking and evidence extraction have relied on simple concatenation of claim and document word embeddings as an input to claim driven attention weight computation. This is done so as to extract salient words and sentences from the documents that help establish the correctness of the claim. However, this design of claim-driven attention fails to capture the contextual information in documents properly. We improve on the prior art by using improved claim and title guided hierarchical attention to model effective contextual cues. We show the efficacy of our approach on political, healthcare, and environmental datasets.

Portfolio drawdowns are defined informally as the loss with re-spect to the most recent peak. In this book we study the optimiz-ation of portfolios for drawdown risk. We optimize portfolios non-parametrically via linear programs and via general numerical optim-ization for bootstrap samples to achieve robust portfolios. Theoretic-ally, we establish the foundation of a parameteric model to optimize for drawdown risk for geometric Brownian motions.

null

We jointly explain the equity and value premium variations in a model with both short-run (SRR) and long-run (LRR) consumption risk. In our empirical analysis, we find that SRR varies with the business cycle, and it has a substantial predictive power for market excess returns and the value premium—both in-sample and out-of-sample. The LRR component also differs significantly from zero, and value stocks have a larger exposure to both LRR and SRR than growth stocks. To explain these patterns in asset returns, we propose an extended LRR model. The model can be solved using log-linear approximations with economically small errors.

Only satisfied clients generate a sustainable profit for a bank. Therefore, besides an intention to generate profit, the client-advisor must focus on client satisfaction and use soft skills to make the client satisfied. Client satisfaction is highly dependent on choosing the right investment strategy. Therefore, the client-advisor must determine the client’s needs (risk, return, unique circumstances) with the client and propose the right strategy. This strategy forms the bridge between the client and asset management. This paper describes my serious game for training people in this holistic consulting process.

This portfolio analysis provides evidence on the risk-return characteristics of investment strategies that align with the transition toward a climate-resilient economy. Expanding the approach of In et al. (2019) to a global perspective, the firms of eight major stock indices were analyzed for the 2008-2019 period. Firms were assigned to portfolios by carbon efficiency, which is defined as revenue-adjusted CO2 equivalent emissions. The results indicate that an investment strategy of buying carbon-efficient firms and selling carbon-inefficient firms generates statistically significant alpha in some markets. Also, a long-only investment strategy of purchasing carbon-efficient stocks does not reduce risk-adjusted returns significantly.

This paper investigates whether an investor can achieve excess returns by applying a momentum strategy to structured products. We compute returns for capital guaranteed products, discount certificates and outperformance certificates for five different maturities and with the SMI, DAX and Euro Stoxx 50 as underlyings. We find that a portfolio consisting solely of prior loser products would have achieved the highest portfolio gains. We also find evidence for a short-term momentum effect and a long-term return reversal. These findings complement previous work as the likes of Jegadeesh and Titman (1993) or De Bondt and Thaler (1985), but stand in contrast to the conception of efficient markets.

null

This thesis examines the equity duration or interest rate sensitivity of aggregate stock markets in the United States, Germany and Switzerland in the current macroeconomic environment of ultra-low (real) interest rates. Using a sample of broad stock market indexes and long-term government bonds, stock returns are found to be positively correlated with the interest rate counter economic expectations. The interacting equity risk premium during this time is suggested to dampen the effect of the interest rate on the equity prices leading to the negative equity duration. Consequently, pension funds should remain reluctant to use equities in interest rate risk management.

This thesis discusses the presence of the low-risk anomaly in the South Korean market. To create the portfolios, I used the historical components of the KOSPI 50. From the numbers it is visible that the difference in alphas between low-risk portfolios and high-risk portfolios is large over the 2000-2020 period. The results of this empirical analysis indicate that the alphas generated are not statistically significant. These results are then discussed, concluding that analysis should be repeated with a larger sample size to receive a better overview of the South Korean market or that the South Korean market is efficient according to the efficient market hypothesis.

null

I examine the size and value effect from 2002 to 2019 in South Africa. I use the FTSE/JSE All Share Index to represent its stock market. I sort stocks on size (market capitalisation) and on value (book-to-price, cash flow-to-price and earnings-to-price). Then, I construct portfolios based on these sorts to analyse the effects. I find a value effect with all three strategies but only the cash flow-to-price one shows a significant effect. I also find a reverse size effect. The effects decrease in the second part of the sample period. Moreover, the value premium is indistinguishable between different firm sizes.

Anfangs der 90er Jahre wurde eine Anlagestrategie durch Michael O'Higgins in den USA sehr populär: Die "Dogs of the Dow" (DoD). Mittels einer Datenanalyse soll für den Schweizer Aktienmarkt anhand des SMI überprüft werden, ob mit dieser Strategie in den Jahren 1994 bis 2018 der Markt geschlagen hätte werden können. Die Renditen des SMI und der "Dogs of the SMI" (DoS) Strategie sind sehr ähnlich und es ist keine markante Überrendite der DoS-Strategie gegenüber dem SMI zu verzeichnen. Eine Investition nach der DoS-Strategie erscheint nicht lohnenswert. Für eine stärkere Aussagekraft sollten jedoch weitere Risikomasse berechnet und Adjustierungen vorgenommen werden.

Das neuartige Coronavirus sorgte im Jahr 2020 für eine Pandemie, die drastische Massnahmen seitens der Regierung und der Bevölkerung erforderte. Dies führte unter anderem zu starken Börsenschwan-kungen, gegen welche sich ein Anleger durch ein breit diversi˝ziertes Portfolio schützten möchte. Diese Arbeit untersuchte am Beispiel der Schweiz, welche Brancheneigenschaften während einer Pandemie den grössten Ein˛uss auf den Diversi˝kationse˙ekt und das Rendite-Risiko Verhältnis hatten. Hierzu wurden verschiedene Portfolio-Simulationen mit gleichgewichteten Branchenindizes durchgeführt und relevante Kennzahlen berechnet. Die Ergebnisse zeigten, dass ein defensives Portfolio keine grössere Risikominimierung als ein zyklisches aufwies. Das attraktivste Rendite-Risiko Verhältnis wurde bei der Gesundheitsbranche festgestellt.

null